DSXchange

Kim, You lost me on how to use this. I've created a routine with your code, now what? Looks interesting. -Craig This routine will ask for a id to a DS_METADATA record. The id is category : "\" : tablename like Hashed\AdSales\Daypart2StgHash. If you compile and test this routine then type in somethi...

This is how you should use metadata to help automate simple tasks. Now this creates an accurate control file. It is simple to use too. Metadata if used properly can help design the best solutions. I think something like MetaStage is overkill for most of us. Metadata should drive your designs from st...

This is in the same order as your metadata in a link if you save a link or if you import. I suggest you create a type 19 hash file. You save all your metadata in a specific category. Select this category call this routine and write the results into this hash file. A type 19 hash file is a directory ...

* ----------------------------------------------------------------- * KgdCreateBcpMetadata(DsMetadataId) * Decription: Create Bcp file from records in DS_METADATA. * Written by: Kim Duke * Notes: * ----------------------------------------------------------------- * $INCLUDE DSINCLUDE JOBCON...

rajainder I posted a few jobs and that will post all the statuses of jobs in sequence. It is on ADN and on my tips page. It is called EtlStats.zip. It should do the trick. You may need to change something to find the failed job within that sequence. If you get those jobs working and cannot figure ou...

If you have one working please post it with code blocks around it so it does not lose anything. I do not mind helping you but you need to do the research. Post one and I can get that routine to do the rest. It should only take a few minutes.

You do not need the batch job when you use the routine posted above. Your first job could read from a sequence job and write to this hash file. Next is the above routine activity. You could call it 15 times or modify it to send it 15 paramaters or hard code the parameter names. Next have all your jo...

Craig is correct. You could do it by backing up the crontab entries ans restoring them. That is only on the UNIX side. You still have at commands that you may miss.

There is no real limit as far as the number of lookups. I have seen jobs with hundreds. I do not recommend that but 30 is not a problem.

If you saved these into DS_METADATA then I would modify my routine KgdCreateDdlMetadata. It will create the create table sql for records in the DataStage metadata hash file DS_METADATA. It should be easy to change this to build what you want. If you post an example of the control file then I might d...

I think you could do it in hash files. It is fairly easy to get a list of all the jobs in a sequence. If you had a 2 part key in a hash file with the job name and the start time then you sort this by time and you got your jobs in order of start time. If you need it by finish time then switch start t...

Ray

Why couldn't you do this in UV.LOGIN in the UV account?

Ken I friend of mine Peter Lai invented this routine which you can call as routine activity and from that point on it will change the value of one parameter. I have heavily modified it. Peter's original version would read all the parameters in a sequential file and set any or all of them. I just wan...

I agree with Ken but if I was going to do this then I would create the project first. Next I would overlay the project with the one from the other machine. This would update the necessary files in the UV account.

No. There is a direct releationship between that one parameter NOFILES and MFILES. It looks like NOFILES is large enough. Just need to increase MFILES.